# Multitask Language Understanding on MMMLU Korean 1.0 (test) Source: [Multitask Language Understanding on MMMLU Korean 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:32.188Z Evaluation of model performance on the test split of the MMMLU Korean 1.0 benchmark for Multitask Language Understanding, reporting Accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Multitask Language Understanding | | Dataset (including any stated split) | MMMLU Korean 1.0 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | CLO | 41.94 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 39.7 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 36.8 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 35.9 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 34.39 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 32.73 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 29.61 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 29.09 | Not recorded | 68fa78ac1e209131cea79c93 | | CLO | 28.31 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 28 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 27.65 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 27.48 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 26.79 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT+DPO | 26.77 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 25.94 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT-tgt | 25.31 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 25.31 | Not recorded | 68fa78ac1e209131cea79c93 | | SFT | 23.47 | Not recorded | 68fa78ac1e209131cea79c93 | ### CLO Accuracy: 41.94. Result ID: 69959b4a3badfaffe696f61a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-3B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### CLO Accuracy: 39.7. Result ID: 69959b4a3badfaffe696f610. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 36.8. Result ID: 69959b4a3badfaffe696f60d. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT Accuracy: 35.9. Result ID: 69959b4a3badfaffe696f619. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-3B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT Accuracy: 34.39. Result ID: 69959b4a3badfaffe696f60e. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### CLO Accuracy: 32.73. Result ID: 69959b4a3badfaffe696f614. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 29.61. Result ID: 69959b4a3badfaffe696f611. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### CLO Accuracy: 29.09. Result ID: 69959b4a3badfaffe696f60c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### CLO Accuracy: 28.31. Result ID: 69959b4a3badfaffe696f618. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 28. Result ID: 69959b4a3badfaffe696f60b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 27.65. Result ID: 69959b4a3badfaffe696f615. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 27.48. Result ID: 69959b4a3badfaffe696f613. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 26.79. Result ID: 69959b4a3badfaffe696f60f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-13B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT+DPO Accuracy: 26.77. Result ID: 69959b4a3badfaffe696f617. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT Accuracy: 25.94. Result ID: 69959b4a3badfaffe696f616. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Mistral-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT-tgt Accuracy: 25.31. Result ID: 69959b4a3badfaffe696f609. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"Target language only"}. This record does not assert independent reproduction. ### SFT Accuracy: 25.31. Result ID: 69959b4a3badfaffe696f612. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-3-8B","Training Data":"English + Korean"}. This record does not assert independent reproduction. ### SFT Accuracy: 23.47. Result ID: 69959b4a3badfaffe696f60a. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama-2-7B","Training Data":"English + Korean"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.